Whale behavior based rule optimization on paraconsistent neutrosophic classification model for prediction of Dyslexia

Abstract

With the advancement in the software field, diagnosing dyslexia in earlier stages among children is highly possible. It helps them to take necessary measures to rise above the problem. This paper intends to develop an uncertainty handling model using neutrosophic logic inference system. This system’s functionality is enhanced by introducing paraconsistent logic with whale behavior based optimization. Paraconsistent logic is used to discover the degree of certainty and contradiction of generated rules. Pruning the population of rules is handled by a nature inspire algorithm known as whale behavior based rule optimization. Dyslexia dataset consists of both vague and crisp values. Treating them as such will often lead to high false alarms in the detection process. To overcome this issue the neutrosophic model is used to denote them in terms of membership degree of truthiness, indeterminacy, and falsity. The paraconsistent analyzer works with the favorable and unfavorable degree of evidence of each rule to handle the inconsistency and uncertainty among dyslexia detection. The potential rules are selected by the encircle prey model of the whale optimization algorithm. The simulation results proved that the performance of the proposed model produces high detection rate in the detection of dyslexia.